Our preprocessing results in round 1 are still available in round 2?

Created by Chao Huang chaohuang
You must 'share' access to your Docker repositories with the challenge infrastructure. The simplest way to do this is to visit https://www.synapse.org/#!Synapse:syn7272737/wiki/ , click `Share` (upper right) and enter `dmchallengeit` and save. After doing this, please submit the file syn8037851 again.
Hi Bruce, Thank you very much for your help! I have modified my submission txt file as you instructed. I received emails saying the submission is invalid. The error message is about the training repository. It says the repository for the image does not exist or is not accessible by the submission processing infrastructure. I have double checked the sha256 sequence and the name of the docker repository. They are all correct. What do you think may possibly be the reason for this invalid submission? Again, thank you very much for your kind support! Best, Yang P.S. The error message is quoted below for your reference. Dear BIG-S2: Your Submission to the Digital Mammography challenge, syn8037851 (ID 8086016), was invalid. The message is: The repository for image docker.synapse.org/syn7272737/dm_feature_tf_2@sha256:6c1dc6f0a075103420685cd411576d2b8982043907004ecaf5f8f21b136035d4 does not exist or is not accessible by the submission processing infrastructure. Please direct any questions to the challenge forum, https://www.synapse.org/#!Synapse:syn4224222/discussion. Sincerely, Challenge Administration
You have preprocessed data stored on the system. It was computed by running this Docker image: docker.synapse.org/syn7272737/dm_preprocessing_r@sha256:d7d3eca888ea6a0048d64271dd6cc6abe323e7654c63c1dee392ae0794a98866 the submission was 8040642 and it ran on or about 2017/01/18 13:09:51, over the course of about 60 hours.   Looking at your Round 2 submissions:   8040601: This was a single Docker repository (not a preprocessing / training pair), so it was taken to be a 'training-only' submission: ``` docker.synapse.org/syn7272737/dm_feature_tf_2 ``` Note: Since you specified no preprocessing, the system understands that you intend to use the raw data directly and does not mount the `/preprocessedData` folder.   8040642: Again, this is a single Docker repository: ``` docker.synapse.org/syn7272737/dm_feature_tf ``` Again, you are telling the system that you do not intend to use preprocessed data. (If that's not clear it would explain the misunderstanding.)   8071642: This time you submitted the following file: ``` preprocessing=docker.synapse.org/syn7272737/dm_preprocessing_r@sha256:d7d3eca888ea6a0048d64271dd6cc6abe323e7654c63c1dee392ae0794a98866 ``` This is a 'preprocessing only' submission. No training occurs.   To specify a training submission that uses this preprocessing data you must submit a file that looks like this: ``` preprocessing=docker.synapse.org/syn7272737/dm_preprocessing_r@sha256:d7d3eca888ea6a0048d64271dd6cc6abe323e7654c63c1dee392ae0794a98866 training=docker.synapse.org/syn7272737/dm_feature_tf@sha256:2d0fa6d042e280a0b6f30d1eb798836e9ba0c2c9bc3fd3109e9f467474eec492 ``` Please see the 'Training submission' section on this page: https://www.synapse.org/#!Synapse:syn4224222/wiki/401759
@chaohuang: I think there is a misunderstanding about how you are supposed to specify your submission. I will have a look at what you have done and comment further. I believe this is your history of submissions: ${leaderboard?path=%2Fevaluation%2Fsubmission%2Fquery%3Fquery%3Dselect%2B%2A%2Bfrom%2Bevaluation%5F7213944%2Bwhere%2BSUBMITTER%253D%253D%25223343190%2522%2B&paging=true&queryTableResults=true&showIfLoggedInOnly=false&pageSize=100&showRowNumber=false&jsonResultsKeyName=rows&columnConfig0=none%2CSubmission ID%2CobjectId%3B%2CDESC&columnConfig1=none%2CStatus%2Cstatus%3B%2CNONE&columnConfig2=none%2CStatus Detail%2CSTATUS%5FDESCRIPTION%3B%2CNONE&columnConfig3=none%2CTraining Quota Remaining%2CTIME%5FREMAINING%5FDISPLAY%3B%2CNONE&columnConfig4=cancelcontrol%2CCancel%2CcancelControl%3B%2CNONE&columnConfig5=epochdate%2CSubmitted On%2CcreatedOn%3B%2CNONE&columnConfig6=epochdate%2CLast Updated%2CmodifiedOn%3B%2CNONE&columnConfig7=synapseid%2CSubmitted Repository or File%2CentityId%3B%2CNONE&columnConfig8=none%2CFile Version%2CversionNumber%3B%2CNONE&columnConfig9=synapseid%2CLog Folder%2CSUBMISSION%5FFOLDER%3B%2CNONE&columnConfig10=none%2CSubmitting User or Team%2CSUBMITTER%3B%2CNONE&columnConfig11=synapseid%2CModel State File%2CMODEL%5FSTATE%5FENTITY%5FID%3B%2CNONE&columnConfig12=none%2CDaily Data Used %2528MB%2529%2CDAILY%5FQUOTA%5FUSED%5FMB%3B%2CNONE&columnConfig13=none%2CDaily Data Remaining %2528MB%2529%2CDAILY%5FQUOTA%5FREMAINING%5FMB%3B%2CNONE}
It's really unbelievable. Since there is no any update for this question, we decided to rerun our preprocessing job. However, I got the email that "an identical pre-processing step had been completed in a previous submission. Therefore no computation was performed for the submission." In that case, we can't have our preprocessed results. How could we do the follow prediction step??
Actually, we tried the version without the dot, and the error continues: STDERR: with open("/preprocessedData/labels/labels.txt", "r") as labels_txt: STDERR: IOError: [Errno 2] No such file or directory: '/preprocessedData/labels/labels.txt' Yang
Hi Yang, Do you intend to start your path in "with open("./preprocessedData/labels/labels.txt", "r") as labels_txt:" with a dot "."?
Hi Bruce, It is not the submission 7837056 that popped out error message. It is the deep learning training submission afterwards. Part of the error message is shown below: STDERR: with open("./preprocessedData/labels/labels.txt", "r") as labels_txt: STDERR: IOError: [Errno 2] No such file or directory: './preprocessedData/labels/labels.txt' The './preprocessedData/labels/labels.txt' is the file output by the previous preprocessing step. The directory is the one we set in the preprocessing step so we are sure that it is correct. What do you think could be the reason for this? Thanks, Yang
Hi Yang and Chao, The pre-processed data from Round 1 were available at the beginning of Round 2. However, please remind that the persistence of the pre-processed data is not guaranteed and depends on the available resources (see first paragraph from the section [Pre-processing submission](https://www.synapse.org/#!Synapse:syn4224222/wiki/401759)). > but it kept popping out error message saying that the preprocessed data are not available. Could you please help us check if those files are actually there or not? @brucehoff Can you have a look at the submission 7837056 to see what is the cause of the reported error? Thanks!
Hi, we tried to train our deep learning model based on the preprocessed results in round 1 but it kept popping out error message saying that the files generated in round 1 can not be found. Could you please help us with this issue? We don't want to rerun the same thing in round 2.
Hi Thomas, I am on the same team with Chao Huang. We tried to train our deep learning model based on the preprocessed results in round 1 but it kept popping out error message saying that the preprocessed data are not available. Could you please help us check if those files are actually there or not? For your reference, the previous job ID for preprocessing is 7837056. Thank you in advance! Best, Yang
Yes

preprocessing results in round 1 still available in round 2? page is loading…